OCCUPATIONAL LICENSING VERSUS COMPANY-LED TRAINING: The controversy over the competence assurance system for European aircraft technicians

The paper provides an analysis of the major controversy which occurred between national aviation authorities during their work on the European harmonization of the aircraft technicians’ competence. The debate focused on the institutional method to assure the high skills level required for that profession: should Europe introduce a system of personnel licensing or a system of company-led training? Should competence assurance and monitoring be assigned to the aviation authority or should these responsibilities be delegated to approved companies? A series of 20 extensive interviews with aviation stakeholders in France, Germany and the United Kingdom showed: (i) the company training option highlights the fact that ‘organization’ is a traditional strategy for competence assurance in high-reliability contexts (like aircraft maintenance), the other main alternatives being occupational regulation and sorting by well-informed markets. (ii) The quality recognition required for the international outsourcing of safety-critical services favours the licensing system. In contrast to licensing, the design of the company training system fails to generate the transparency and trust necessary to reassure foreign customers. (iii) Both systems of competence assurance are densely intertwined with specific industrial relation patterns and vested interests. The distortion of these – national or local – equilibriums by European harmonization encounters strong opposition from the industry and/or unions.

The European states’ aviation authorities have achieved the impossible: they have succeeded in completely harmonizing qualifications of the vocational type and having them automatically recognized across Europe. These qualifications consist of training syllabuses and licences for mechanics and technicians in aircraft maintenance. After seven years of discussions and a further seven-year implementation phase, the licensing scheme was introduced in Europe in October 2006. The national aviation authorities are responsible for managing and monitoring the system in their respective states, whereas the newly established supervisory authority, the European Aviation Safety Agency (EASA), is in charge of drawing up regulations and supervising their implementation in Europe.

The modular structure of the training content and the corresponding examinations makes the framework quite adaptable. The modules can be combined to form ab-initio courses, as well as further educational programmes, re-training measures or self-improver schemes. Full-time initial training courses for mechanics take one year and courses for technicians take two years. Additional practical aircraft maintenance experience (at least one year for mechanics and two years for technicians) is also required to qualify for the respective licenses. We are not aware of any other standardized European qualifications below higher education level that demands such a long period of training.

The harmonization of aircraft maintenance qualifications is in line with current European policies on cooperation in educational matters. The move towards cooperation, accelerated since the 1990s, is encompassing the fields of regulated professions as well as higher education (the ‘Bologna process’) and vocational training (the ‘Bruges-Copenhagen process’). The maintenance project is part of this cooperation because it makes provision for simplified cross-frontier recognition of certifications. This goal, which is strongly supported by the European Commission, is expected to give European citizens several advantages: freedom of establishment, free provision of services, and smooth cross-frontier educational and labor market mobility.

At the same time, the standardization of qualifications which has occurred in the field of aircraft maintenance – and the transfer of the responsibility for rulemaking to a supranational body – is an exceptional case in Europe. Although the European States have managed to develop fairly uniform technical standards, economic regulations and social laws, they have not harmonized their systems of training and qualifications. The efforts made by the European Commission on these lines eventually brought to light the impossibility of harmonization. They failed owing to the strong resistance put up by the nation-states and to the lack of market-based demand (Orzack 1983; Bjornavold and Sellin 1997; Berggreen-Merkel 1999). The nation-states have consistently refused to make any concessions when it comes to educational sovereignty; for example, in the European Treaty of 1992 the states laid down paragraphs requiring programmes and actions by the European institutions to be in accordance with the principle of subsidiarity, excluding any drafting and designing of common training policy and prohibiting any standardization of the regulations, contents and certifications in the field of education. On the market side, there has been no demand for simplified recognition of qualifications because interest in cross-border mobility is very low in general. And there was also a lack of interest on the part of the companies – the idea of reducing the cost of foreign investment by harmonizing training has not yet found any support worth mentioning (Rolfe 2001; Severing 2005).

The initiators and negotiators of the maintenance project, the European aviation authorities, nevertheless agreed on harmonization without any great disputes. The reasons why it was so easy to reach a consensus on the standardization of the range, content and depth of the maintenance training are discussed in the first part of this paper.

The fact that it then took seven years to discuss and prepare the regulations was due to the large number of controversies over the institutional precautions for assuring the agreed competence profiles. This paper focuses on a topic of debate that continues to resurge even today. The debate turned on the question as to whether a licensing system or a company training system should be introduced. The result was that the licence option was chosen. This solution and the path towards it are of interest from both the sociology of the professions viewpoint and in terms of international educational cooperation. These aspects are analysed in the second section.

The third section presents some of the study's conclusions. Our results support the ‘domestic issue linkage’ concept of Falkner et al. (2005), for example. According to this, a European political project can generate unintended side-effects that come into conflict with domestic processes in other political fields. The French aircraft maintenance case provides ample illustrations of this link. Because of the many interconnections between the competence management model and specific patterns of industrial relations, the changes introduced by European efforts towards harmonization have encountered strong opposition in France at both employer and trade union levels.

The data for the study was mainly gathered during 20 semi-structured interviews with experts from Germany, France and the United Kingdom.¹ When the European maintenance qualification was created in the 1990s, the interviewees were either acting as decision-makers or privileged observers. The study was financed by the French Ministry for Education.²

One of the key driving forces behind the standardization project was the companies’ real need for uniform skill profiles. The project therefore did not suffer from the common lack of interest in the European institutions’ educational initiatives. A factor that was particularly helpful in realizing the project was the limiting of the standardization to a special function group with an internationally similar competence profile (certifying staff). This avoided the usual conflicts regarding the outline of the profession and training content. Furthermore, the safety regulations for the maintenance sector forced the training to be established at a homogeneously high level. This avoided disputes about the level and interpretation of the standards, which commonly occur in other projects.

The fact that skills and qualifications have been standardized in the area of aircraft maintenance is thanks to the strongly growing need for the outsourcing of safety-critical services. Since the 1980s, many national airlines have developed a strong interest in transferring maintenance tasks to third-party companies in other European countries, owing to capacity and cost reasons. However, this outsourcing has so far been massively hindered by Europe's fragmentation into national maintenance regimes. Because of its particular potential for danger, aircraft maintenance may only be carried out by organizations that are regulated and recognized by the national supervisory authority. By outsourcing maintenance tasks to other European countries, the aviation authorities of the outsourcing state were forced to approve the companies in those other countries. Acquiring this approval was expensive, complicated and long-winded for everyone involved. The maintenance companies abroad had to adopt the special national regulations of the outsourcing state. In addition, the outsourcing company and its national supervisory authority had to make regular onsite checks to ensure that the regulations were adhered to.

In light of the increasing demand for a European maintenance area, the fragmentation into national regimes was phased out in stages during a lengthy, 25-year process. An initial phase tackled the technical guidelines of aircraft construction and maintenance. This was followed by regulations for the internal organization of maintenance operators. At the beginning of the 1990s, work finally began on the component that was still needed to achieve full mutual recognition of maintenance services – the standardization of the maintenance personnel competence profile.

The negotiators quickly agreed that harmonizing the competence profile of a specific, strategic subgroup would be sufficient to recognize the workforce of a maintenance company as being qualified. The target group in question is the certifying staff. According to estimates, there are at least 50,000 persons belonging to this group in Europe. They constitute around one-third of the total maintenance personnel. The remaining two-thirds are mainly technical personnel who carry out maintenance tasks under supervision.

Certifying staff are mechanics and technicians to whom the company has assigned the role of certifying the correct state of the maintained technical systems. With their signatures:

• -

  they certify that the aircraft or part complies with the airworthiness requirements in force;

• -

  they certify that the aircraft or part is in a condition for safe operation; and

• -

  and they release the aircraft or part for return to service.

The certifier will be required to have the European license. Release authorization requiring a licence for the first time in some states, the certifying staff will thus become a regulated, access-restricted occupation across Europe. The licence is portable, but the release authorization is not; when staff change companies, the authorization is lost and can only be issued by the new company after a familiarization period of several months.

The naming of personnel with authority to release to service and the quality statement by signature are two institutions that apply worldwide. They reflect the fact that aircraft maintenance is rated as an area of high potential danger. Maintenance thus needs to be incorporated into a high reliability organization (HRO) whose precautions protect company-internal and external customers from risks and disasters. The precautions of an HRO include the two institutions mentioned previously. As well as setting up a competent gatekeeper, they ensure that responsibility can be retraced and attributed to specific persons.

To fulfil their assignment, the certifying staff are often given the dual role of technical specialists and inspectors. As technical specialists, they have the exclusive right to make safety-critical interventions themselves (trouble-shooting) and to monitor routine tasks; as inspectors, they have the exclusive competency for signing-off the system once the checks have been completed. The previously widespread existence of this dual role in the world norm is, without doubt, a key factor in facilitating the skills standardization. The wide distribution enabled a consensus to be reached quickly on the range of competences. The competences were to include both those of a technical specialist (systems, experience and skills) as well as those of an inspector (system interdependencies, quality and regulations).

A rapid agreement was also reached on the syllabus. This can be explained by the existence of established reference lists. These registers found general recognition owing to their quality and neutrality. Those that should be mentioned are the skills catalogues of the UN agency ICAO (International Civil Aviation Organization) as well as the association of US airlines ATA (Air Transport Association). The implicit textbook function of the maintenance handbooks issued by major manufacturers should also be mentioned. The manuals from Airbus, Boeing, Rockwell, Rolls-Royce, etc., inform the international maintenance community of ‘… what good certifying staff need to know and be able to do’ (quotation from the interviews).

A special feature of the harmonization project was the structural obligation to establish training at a homogeneously high level. The failure rate of more than 70 percent in some countries during the first examination years shows that the European partners have designed an extremely high standard with the common licence.

There is a general assumption that non-standard or moderate minimum standards would have failed in the goal of creating an effective internal European market for safety-critical maintenance services. In an area with high quality demands, only homogeneously high skills standards can build up the trust required for the recognition of foreign services. The European partners had the negative example of the ICAO licence model in mind. This was created in the 1950s and has been updated ever since. It was supposed to become a globally recognized reference model for maintenance personnel qualifications (ICAO 1957). However, national practices are so full of variants in terms of completeness and level of ICAO training that the resolution of mutual recognition of the ICAO-compatible licences failed.

The inquiry on debates which eventually took place during the European works on maintenance qualifications revealed one overarching controversy. This dispute focused on the institutional method to assure the high skills level required: should Europe introduce a system of personnel licensing or a system of company-led training? Should certifying staff, hence, hold a licence or a company authorization?

Company training or individual licence – this debate is by no means new in the maintenance field. A historical example is the ICAO licence project mentioned. This project planned to give states the choice between a licensing system and a system of company training. The express acceptance that both systems were to be viewed as equally valid provoked controversy even back then. While such diverse countries as Australia, Finland and Venezuela expressly rejected the company training option, other countries such as Afghanistan, Ethiopia, France and the United States had introduced this system and insisted on its recognition (ICAO 1958).

One factor that the two systems have in common is that the company names the certifying staff. In a licensing system, it is the state (the national aviation authority) that satisfies itself that the candidate possesses the necessary competence. In a typical case, the authority requires the candidate to attend a series of accredited training courses, offer proof of a qualified work history and pass a demanding examination. In return, a licence is issued that serves as both a proof of basic competence and a permit.

Under a system of company training, by contrast, the competence assurance is delegated to officially accredited maintenance companies. The role of the supervisory authority is restricted to approving and making spot checks of the internal organization at the firm. Typical precautions taken by the firm to ensure competence include (a) establishing an internal quality department; (b) constant evaluation of personnel and training results using a technical and hierarchical superstructure; and (c) determining and budgeting for a programme of initial and further training. The latter includes the highly selective recruitment of young graduates of vocational schools, the training of these new recruits in an internal course and testing system and the assignment of these new recruits to in-house careers with an integrated professionalization plan (training on and off the job). A waiting list of qualified and experienced candidates is created; the company uses this to determine the certifying staff. Unlike in the licensing system, the company does not require approval to authorize a person. A career in an accredited training system of the company is sufficient as proof of competence. There is no licence in this regime of internal training and job markets.

The model of company training offers those involved some advantages in comparison to the licensing system. The supervisory authority has the advantage of passing on monitoring costs to the companies (testing, traceability of work histories, record keeping and release authorization checks). And, as will be demonstrated below, the companies have more leeway: company-specific instead of standardized skills profiles, allowances in assigning release authorization, and moderation in wage levels.

Twenty-five of the 27 states involved in the European project already had a national licensing system for maintenance personnel. But the systems were so divergent that cross-border recognition was not granted. In addition, the countries often had a mixed structure of licences and company training systems. The latter were conceded to major airlines as a special dispensation. Only two countries decided historically on a system of company training: Belgium and France.

The controversy on the institutional method to assure the commonly agreed skills profile split the European maintenance community into two camps with France on the one side and the rest of Europe on the other. France pleaded for both the licence and the company systems to be given equal value and demanded a free choice of options for each country. It called on the subsidiarity principle as well as the statistically proven fact that both systems possess an equally favourable safety standard. It was clear that France would choose the option of company training. This variant was rejected by the rest of Europe. The countries demanded an exclusive licensing system. This, in turn, was subject to great opposition from France.

The following passages present an analysis of the points of view of the main actors involved. It makes sense to divide the French opposition into trade unions and company management.

Each of France's five major trade union confederations, all represented in the maintenance sector, was very reluctant when it came to the licensing idea. The strongest union in this area, the CGT, strictly rejected a licence. The rejection of the French trade unions is surprising. It goes against proclamations by the international association of transport workers’ trade unions ITF which had spoken out clearly in favour of a licensing system and against the company system. The rejection is also astonishing in view of the fact that a licensing system typically brings wage benefits with it. So why did a trade union like the CGT still reject the European licence?

The experts who were interviewed often tried to give ‘politically correct’ answers to these questions. The research therefore only allows us to suggest a possible reason based on the interview evidence. The rejection of the licence seemed to be linked to the organization's interest in protecting its survival and influence. One interview respondent explained as follows:

This driving out can be seen in all fields of French aviation where a licence is required. The interests of pilots, cabin crews, navigational engineers and air traffic controllers are represented in the vast majority by small professional associations (‘craft unions’) and not – as in the other areas of aviation industry – by large encompassing confederations (‘general unions’). The underlying process leading to this shift is well-known (cf. Wagner 1966): the licence creates a special group of people who have organizational ability owing to their particular interests and who are extremely willing to organize because of their strategic position. This situation offers political entrepreneurs favourable opportunities to organize, channel in and implement a special interest group.

The correlation between licences and craft associations is a distinct possibility for (general) union protectionist argument. A second possibility lies in the high fragmentation of the French trade union landscape. The major confederations are trade unions that are linked to one ideology or party and that are in ideological conflict. They are surrounded by a circle of particular interest groups. Air France, for example, currently (2006) has a total of 19 craft or special interest associations in addition to the five large confederations.

The union protection argument is also backed up by the CGT's strategy of preventing any chance for occupation-specific policy to make its mark while the licence is being introduced. The representative of this policy, the small craft union for maintenance mechanics (SNMSAC), was not able to exert its influence. By using its dominant position on the committees, the CGT stopped the issue of wage dividends for release authorization from being put on the agenda, for example. It also made sure that not just the certifying staff, but also those on the waiting lists received a licence during the system introduction. Its (successful) resolve to regulate sanctions against licensees by law and not via company agreements is also in line with the exclusion of image building opportunities for craft unions.

Additional support for the union protection argument is that the licensing system builds up a platform for strongly competitive alternative views in an ideological and material field. The competition becomes especially clear when looking at the remuneration for the release authorization. In the case of the company training system, this authorization does not bring special wages with it. General unions, such as the CGT, justify this on the basis of the principle of preventing dualization of staff as high and low-qualified employees. According to their arguments, companies are obliged to train and further train all staff equally well. The release authorization would thus not require any additional skills and would therefore also not permit any special rewards. The licensing system stands in sharp contrast to this. With this system, the authorization results in a clear increase in income. This is legitimized by the constructs of ‘special responsibility’ and ‘individual competence’ of the certifying staff. The competitive strength of this alternative view, and thus the risk to the survival of an egalitarian-oriented trade union in a licensing system, is clearly shown by the relative remuneration position of the best-paid maintenance technician. This is the head of the complete maintenance for a Boeing 747. Under the licensing system at British Airways, he or she earns the same as an experienced head pilot for this extremely complex job. Under the system of company training at Air France, he or she earns 60 percent the salary of a young co-pilot (source: interviews).

The comparatively moderate wages under the system of company training are based on a specific arrangement of training and work organization. To keep costs resulting from the egalitarian professionalization of everyone within limits, the company structures the training – and thus the skills – of individuals in rather small claims. Unlike in the licensing system, maintenance personnel thus frequently specialize in just one or two aircraft components. This legitimizes a moderate positioning of their wage level and leads to an overall flatter grading of wages.

The supervisory authority and maintenance companies in France also led a tough defensive battle against the licensing system. But the central reason for this was not the cost of system transformation. According to our research, the rejection of the licences was based more in the interest of preventing an (additional) disruptive group from settling into the workflow. One interview partner explained the reason as follows:

Air traffic safety regulations specify that a maintenance programme has to be carried out during each stopover. If a licence is required to carry out this intervention, a small group can hit an airline hard or even totally block it with a wildcat strike or walkout. In the system of company training, on the other hand, it is far more difficult for disruptive powers to develop, with industrial action requiring a far broader mobilization. Unlike in the licensing system, companies have reserves that they can recruit quickly to maintain a minimum service. These include the waiting lists of qualified candidates mentioned earlier. In addition, it was also common for middle technical management to be assigned release authorization.

In countries with a strong licensing tradition, such as the United Kingdom and Germany, the potential disruptiveness of maintenance personnel did not cause any concern. So why did it in France? The problem, implied in the quotation above, is the ‘gréviculture’ (Eymond-Lariaz and Javault 1999) that is prevalent in France. The term refers to the widespread use of small, very sudden walkouts. This type of industrial action is generally used by strategically placed groups in the area of state or state-controlled infrastructure (transport, education, healthcare, etc.). The ‘gréviculture’ can be illustrated using the example of Air France. As the privatization of this flag airline drew politically nearer in 1997, there were a total of nine walkouts within the space of 12 months, carried out alternately by seven different employee groups (Bouaziz 1998; Autier et al. 2001). The pilots’ strike achieved world fame during the weeks when the football World Cup was being battled out in France. It is thus understandable why the French aviation industry managers were opposed to the licence. They wanted to prevent the already pronounced ‘gréviculture’ from becoming stronger. This position was strengthened by the intensification of the conflict climate which arose along with the preparations for privatizing the French flag airline.

The pattern of small, sudden walkouts results from a system of institutions that are very different from industrial relations in northern Europe. Three factors that act in favour of the ‘gréviculture’ must be highlighted at this point (cf. Eymond-Lariaz and Javault 1999; Andolfatto and Labbé 2006). These concern the legislation on strikes, corporate co-determination and the securing of trade union survival. In France, there is no law that regulates industrial action. Obligations on employers and employees to avoid conflict and to negotiate and ballot are unknown. Corporate co-determination is restricted to consultation rights; local walkouts are thus, above all, a means of expressing dissatisfaction, pointing out shortcomings and demanding company measures. After all, the French trade unions are not member organizations – they are solely associations of activists. They do not base their organization's survival on selective services for members as the trade unions in northern Europe do – instead, they base it primarily on measures for increasing visibility in the local arena, in other words, on confrontational tactics in the company.

Outside France, a chorus of European aviation authorities, companies and trade unions developed, which spoke out for the licence and energetically against company-organized training. In effect, these stakeholders did not believe that the company system was robust enough to shield training quality in the face of economic pressures. The following quotation from our interviews illustrates this scepticism.

According to Gómez-Ibáñez et al. (1999): 544), maintenance operators are tempted to abuse the technical robustness of modern aircraft systems. The robustness significantly extends the length of time it takes for maintenance deficits to be observed. ‘Some carriers that take little effort … can take advantage by masquerading as high-safety carriers, and charging a premium price. The incentives to engage in this kind of behaviour are strong because the costs of crash prevention are borne in the present, whereas the effects of crashes occur at randomly defined points in the future. Even a carrier that becomes very careless may not suffer a visibly increased crash rate for several years. In the interim, the carrier can earn excess profits’.

Another illustration refers to the effects of business cycles. The aviation sector, and thus the maintenance area, is traditionally shaped by shifts between pronounced periods of activity and slumps. During phases when the economy is on the up, companies must quickly increase their maintenance personnel. This encourages them to ‘cut corners’ – in other words, to cut back on the length of training, thin out professionalization programmes, lower the required level of experience, and not send personnel on training courses because they are indispensable, etc. Note that these measures are not necessarily illegal. However, they do remove the skills surplus that is essential for sustained quality production in HROs. During phases when the economy is in decline, internal company training also comes under pressure. Company training budgets are more flexible than most other areas of expenditure, because of their lower standardization. Furthermore, they count as expenditure that is particularly high, but that does not result in any (positively measurable) profit. In the case of an economic slump, the risk therefore arises that the companies reduce their training expenditure to an inadequate level. This correlation is well-known in aviation. It is the reason why finance observatories have been established within many supervisory authorities, for example. They use these to scrutinize the financial data of the aviation companies. The information is used to publicize the supervision of companies in difficulties, promptly and visibly for everyone.

The large majority of stakeholders outside France opted for the licence because of the instrument's economic independence. The point of view of the international transport workers’ trade union ITF (1997) is representative of this view: ‘Aircraft maintenance technicians should be qualified under a state-issued licensing system with the privileges of the licence vested in the individual, not the maintenance organization. … The absence of requirements for state-issued licensing removes one of the most effective checks against the downward pressures of commercial interests on day-to-day maintenance practices and standards’.

Efforts to protect against economic opportunism can be seen throughout the new European licence framework. Take the rules guiding the centres for aircraft maintenance training. There are precautions for protecting from underinvestment: instructors shall undergo updating training at least every two years relevant to current technology, practical skills and the latest training techniques; the maximum number of students undergoing practical training during any training course shall not exceed 15 per supervisor. Other measures clearly tend to suppress local deviations from the agreed syllabus: only multiple choice tests are to be used in examinations, and the test questions are taken randomly from a central European database of questions.³

The exclusion of the recognition option as an alternative to harmonization is a further illustration of the extremely strong quality assurance efforts. In principle there was no need to harmonize because the European Directive 92/51 EEC, issued in 1992, already provided for the ability of licences throughout EC Member States to be recognized as they would in their own country. The Directive requires the Member States to accept any attestation qualifying a person for his or her chosen profession in his or her state of origin without any further formal examination. It covers mutual recognition of certificates (issued after completion of a vocational course) as well as attestations of competence (awarded upon assessment of work experience), the latter case being indeed applicable to company training systems. The Directive represented a shift from detailed regulation towards a looser framework arrangement, based on mutual trust in the quality of qualifications granted by other states. In the aircraft maintenance case, however, this mutual trust definitely did not exist. One of the national decision-makers interviewed expressed this in the following terms:

The pressure to introduce a harmonized licensing system must therefore also be understood as a means to circumvent legal obligations to recognize certificates issued by unfamiliar or distrusted training systems abroad.

The dispute surrounding the system of competence assurance seemed to be unsolvable. Both sides persisted in their viewpoints. A majority decision was not possible because the assembly of European aviation authorities only made decisions unanimously. France was one of the major centres of aircraft maintenance so it would have been barely conceivable to exclude it from the standard. So why was the licensing system finally accepted?

The system was accepted as part of a political trade-off between the aviation authorities and the European Union. The European Union – the Commission together with the majority of the Council – offered the aviation authorities an agency with which aviation norms could be more effectively standardized. In return, the aviation authorities (including France) conceded to the licensing system, allowing the Union to achieve automatic cross-border recognition of the new certificate.

The aviation authorities’ weak point was the way in which they were organized at European level. They had formed themselves into a type of independent club without any organization worth mentioning. Their agreements required unanimity, and the conditions for the national implementation of norms were dependent on the good will and local power of the respective authority. The mutual checking of the implementation remained rudimentary. The result was sluggish progress, agreements being made only tentatively and standards, especially technical norms, continued to vary noticeably. This was a disadvantage, particularly for Europe's influential aircraft manufacturers concentrated in France. They did not see why they had to build any number of variants of the same aircraft for different countries and/or to invest time and costs for duplicative certification by multiple authorities.

The European Union – or more precisely, the EU Commission – joined the debate as a deus ex machina. It offered to set up a central, well-equipped agency (EASA) to replace the club and to standardize technical norms in the construction and maintenance area more effectively. It linked this offer to the condition that a licensing system be introduced in aircraft maintenance and put under the direction of EASA.

This demand can be explained by the European Commission's constant struggle to establish mutual recognition of educational qualifications. During the second half of the 1990s, the critical phase of the negotiations on maintenance qualifications, the balance of European educational cooperation was rather poor. Until then, the Commission had achieved mutual recognition only for a series of liberal professions and after considerable difficulties (Orzack 1983; Berggreen-Merkel 1999). A new approach, the ‘Bologna process’, was launched by some of the European governments in 1998 with a view to standardizing some of the basic European higher educational institutions, but the Member States of the Union systematically thwarted the Commission's efforts to establish itself as the coordinator of this movement (Croché 2006). Moreover, at vocational training level, the Commission had experienced a chain of defeats. Campaigns mounted since the 1970s for the recognition and equivalence of vocational qualifications had failed (Bjornavold and Sellin 1997). As mentioned above, the European Treaty of 1992 even expressly prohibited the Union from intervening in national training systems. The Directive 92/51 EEC on mutual recognition of certificates was not only rejected, as described, by aircraft maintenance circles, but rather by many publics across Europe (Commission of the European Communities 2000). Some countries certainly developed bilateral agreements on the mutual recognition of vocational qualifications; since the 1980s, for example, Germany has been signing agreements with France on the mutual recognition of 19 qualifications in all, and with Austria on 131 qualifications. However, subsequent modernization of the training by one of the partners put many of these agreements into question (Zedler 2006).

Helping the licence to triumph in the aircraft maintenance area therefore provided the European Commission with a welcome opportunity of scoring a few points for European diploma recognition policy.

The case study on aircraft maintenance leads to several suppositions related to European educational cooperation. First, the company training model existing at a global scale, and even dominating in some countries, causes us to recall organizational ‘self-regulation’ as an established variant of competence assurance. This variant is sometimes overlooked in the current dispute between free market and regulation. Secondly, the quality recognition required for the international outsourcing of safety-critical services (like aircraft maintenance) favors the licensing system. In contrast to personnel licensing, the design of the company training system fails to generate the trust necessary to reassure foreign customers. And thirdly, the French case shows that the established system of competence assurance is densely intertwined with domestic industrial relation patterns and vested interests. The distortion of this arrangement by European harmonization may therefore trigger strong opposition.

Both in the political arena and social sciences, discussions primarily encompass two fundamental institutions for competence assurance: market transparency and occupational licensing (OECD 2000; Law and Kim 2004).

Market transparency serves to produce reliable information about the competence of the providers and thus to increase the incentive to make adequate investments in human capital (OECD 2000). The appropriate methods of creating transparency include, in particular, networking, building up and spreading reputations, as well as public valuation by recognized rating agencies. For some authors (Friedman 1962; Stigler 1971), transparency is sufficient to generate equilibrium at a high competence level in the market. A licensing system is, from this point of view, a disruptive factor for the market and is thus suboptimum. It occurs as a consequence of weaknesses in the information flow or results from the social closure of a group for the purpose of achieving monopoly profits.

The transparency-based market solution is considered poor by other authors (Akerlof 1970; Leland 1979). They refer to the permanent, latent risk of adverse selection: the investments required to ensure competence lead to a high price level; this attracts providers that offer price discounts through lower investment; with lower prices, entry becomes unattractive to new quality providers and existing quality providers quit the market; the quality level sinks and a second class segment or low competence equilibrium arises. Homogeneously high competence equilibrium can only be established under a licensing system according to these authors, the key mechanism here being to secure the full returns for high investment. With its entry regulations, the licensing system prevents ‘low investment – low price’ substitutes from penetrating the market. Entering and staying in the market thus remains attractive to quality providers.

The French maintenance case suggests that we should consider ‘self-regulation’ as a third important variant. A key characteristic of this option is that the state delegates the competence assurance to the companies. These can group themselves together in industry associations or, as in the system of company training, operate company-internal structures under state supervision (see Ayres and Braithwaite 1992). In the latter case, the company quality department is the central agency for competence assurance. According to the regulations, it is completely independent from other departments, assigned immediately below the company director and equipped with a staff of experienced inspectors to ensure that it is effective. The internal quality framework must include a robust feedback mechanism of audit findings to ensure, as necessary, corrective action.

Among the three competing systems of competence assurance, the licence solution seems to be winning through in the course of European harmonization.

No states have regarded the market transparency method as adequate since post-war times. According to our estimate, the distancing from this option is based on the coming together of several structures that prevent market transparency in the maintenance area or that render this ineffective. These structures include incubation (quality deficits only become identifiable after a long time), the identification problem (it is difficult to assign individual blame) and market sensitivity (an accident assails the market presence of the ‘victim’ above all, i.e., the airline).

The system of company training is also on the retreat, despite the advantages for employers in terms of flexibility and wage moderation. The retreat is no longer restricted just to aircraft maintenance – it has also reached many safety-critical functions in the transport sector. There is an impressively long list of European standardization projects where the same substitution programme is on the agenda. Whether it involves pilots, flight operation officers, flight attendants, air traffic controllers, train drivers or merchant navy deck and engineer officers – the existence of company training systems is constantly being threatened and these systems are being increasingly replaced by standard licensing systems.⁴

The paper has given a political and an economic explanation for the substitution process. The political explanation refers to the political trade-off between the European Commission and the supervisory authorities. The authorities’ monitoring and rule enforcement problem gave the Commission a welcome opportunity to make progress at the level of cross-border diploma recognition. The economic explanation highlights the licensing system's advantage of being relatively resistant to market pressures and opportunism.

There is another – social – explanation in addition to these two explanations. This explanation begins with a paradox. Companies that are in a national mixed structure of licences and company training systems treat the company system with suspicion if it is located abroad. If, however, it is in a domestic company, the system is trusted. France is a textbook example of a flourishing domestic market that is structured exclusively on in-house systems.

In our view, the explanation for this paradox lies in the fact that foreign companies are excluded from the domestic professional ‘milieu-like’ community of staff, competitive companies, schools and supervisory authority. This exclusion from the national milieu makes key information about the respectability of a domestic provider inaccessible.⁵ With this argument, we follow Perrow's analysis of the importance of the network factor in the quality control of safety-critical tasks and HRO organizations. ‘Simply put, the more organizations and groups are involved in protection, the less likely the cover-up. If the organizational field is rich, and thick, the cover is hard to pull over the crime … One should have the rich environment of diverse interests that makes a cover-up just too risky in the first place’ (Perrow 1999: 153).

In the course of our investigations, we were made aware of the vitality of this milieu in many places. What the national authorities overlook in their audits circulates in the network as gossip. As one of the UK observers puts it, in describing the community of over 500 aircraft maintenance operators in his country:

The milieus and networks are, however, also sealed off by state, according to our observations, in spite of the internationalization of the transport sector. Foreign companies and authorities are thus barred from the community-based control component. This exclusion might also explain why the foreign parties involved are keen on the licence; its density of rules compensates for the lack of cross-border social density.

In their research into the EU social policy Directives, Falkner et al. (2005) found significant differences between countries’ adherence to these regulations. After examining a variety of explanations for this variance, they favoured the concept of ‘domestic issue linkage’: ‘We have found that national adaptation is frequently linked to other political processes at the domestic level. (…) The sheer number of cases in which issue linkage has played a role indicates that this is far from being a negligible phenomenon. (…) One fundamental lesson to be drawn from our research is that we should always be aware of the possibility that the “downloading” process of EU policies may become intertwined with the idiosyncratic logics of domestic policy-making’ (Falkner et al. 2005: 313–4).

In our study, France is an excellent example of this concept. The strong rejection of the licence by the main French parties involved was not a result of inherent deficits in the licensing system; the qualities of the licence as a means of competence assurance were not questioned. But the project was for a period strongly rejected because it (unintentionally) came into conflict with domestic politics in a different field. In our case, these were political strategies in the field of industrial relations. The licence – a project solely concerned with qualification policies – threatened to thwart projects of power politics. From the point of view of the major trade unions, the license weakened the association's presence and influence in a context of interorganizational rivalry. Even the management of the aviation companies did not reject the licence because of quality reasons, but because of power-political reasons, as described. They wanted to prevent another strategic group from gaining access to the ‘gréviculture’ instruments.

Educational policy is, without doubt, at a crossroads from which many domestic issue linkages are possible. To us, it seems to be a sensible hypothesis to look for the explanation of the repeated failure of European educational policies among the opposition resulting from this linkage.

Joachim Haas is a researcher at the CEREQ Institute, Toulouse unit (Centre d’études et de recherches sur l'emploi et les qualifications). Studied sociology and political science at Konstanz, Paris and Heidelberg Universities. Research topics: educational and employment systems in a comparative perspective, industrial sociology of aerospace engineering and its supply chain.